Why AI in video conferencing?

TrueConf AI Server software is an AI-based (machine learning) solution for speech recognition in conferences from TrueConf. It automatically recognizes the audio from your events conducted on TrueConf Server and saves it in text format with speaker separation. It can recognize speech from both group conferences and point-to-point video calls (which are also converted into conferences).

Thanks to seamless integration with TrueConf Server and TrueConf Enterprise solutions, users can access protocols through their video conferencing server without needing to log in to a separate service. This enhances user convenience and reduces the number of different web pages users need to access. We recommend using the video conferencing server’s personal area to manage protocols, although it is still possible to log in via the personal area of TrueConf AI Server.

The solution is provided as a software for installation on the customer's premises (on-premise version).

Hereinafter, the terms decoding, transcription, and speech recognition are synonymous and refer to the process of analyzing an audio conference to obtain its transcript (script).

Main Features

• Automatic speaker detection for accurate attribution of each statement.

• Generating a complete transcript of the conference with phrase separation for each participant (including terminals in meeting rooms).

• The ability to select the primary language for audio track recognition in a conference.

• Display of timestamps (relative time) for individual phrases in the transcript.

• Automatic punctuation insertion.

• Automatic capitalization of the first letter in sentences.

• Correct speech recognition when speakers use different languages.

• Accurate recognition not only of commonly used speech words but also of those that are rarely used.

• Recognition of known abbreviations.

• Correct recognition of numbers and dates and recording them in digital format.

• Successful recognition of audio tracks in challenging acoustic conditions: noise, soft speech, simultaneous speech from multiple participants.

• The ability to recognize audio on both NVIDIA GPU and x86-64 architecture CPU. TrueConf strongly recommends using the discrete GPU option for better recognition performance.

• Support for multi-processor systems when processing audio on the CPU.

• Storage of conference audio recordings and speech-to-text transcripts.

• Manually delete any conference recording, including its transcript.

• No limitations on the number of recognized conferences thanks to task queue creation.

• Support for audio processing on systems with both single and multiple NVIDIA GPUs.

• Web interface for managing recognized conferences.

• Authorization on the AI server using a video conferencing server user account, including when the customer's side uses an LDAP directory service, such as Active Directory.

• The ability to integrate the AI server with one or more video conferencing servers.

• Display of ongoing conferences' information: name, Conference ID, start time, and duration.

• Built-in web interface player for playing back the audio file of a conference or call.

• Ability to configure decryption access for both conference participants and any arbitrary user of the video conferencing server.

• Real-time display of conference recognition status.

• Export decryption results in text format (.txt) or in tabular format (.csv).

• Export the conference's mixed audio track.

• Notification to the conference moderator about the successful saving of its recording on the transcription server.

• Conference moderator notification of the successful completion of the transcription process and providing a link to the result.

• Notification to the videoconferencing server users that they have been granted access to view the conference transcription.

Operating Principle

To use TrueConf AI Server, TrueConf Server version 5.5.0 or higher is required. Additionally, the integration capability with the AI server is governed by a separate license for TrueConf Server:

The ability to integrate the AI server with one or more video conferencing servers.

• Events related to pre-recording on the TrueConf Server side.

• your TrueConf AI Server license, which controls the number of video conferencing servers that can be connected to the AI, as well as depends on the capacity of the purchased hardware and software package.

The following algorithm is used:

1. Install TrueConf Server and TrueConf AI Server on different machines.

2. TrueConf AI Server is installed on a dedicated physical machine (virtual machines are not supported as AI operation requires maximum interaction with the hardware).

3. The video conferencing server is connected to the AI server using an integration key. In this setup, multiple TrueConf Server can be connected to a single instance of the AI server.

4. According to the settings of TrueConf Server (whether all events or specific ones are logged), the audio recording of the conference is sent to TrueConf AI Server.

5. Speech recognition and transcript creation are in progress. This can be done automatically right after the event ends or manually initiated by the conference owner in the personal area of the AI server. Additionally, the owner can share the original audio and the resulting text with any participant and even other users of their TrueConf Server.